Update app.py

app.py

@@ -5,35 +5,33 @@ import requests
 from PIL import Image
 from io import BytesIO
 # Import necessary libraries
-import torch
 from transformers import pipeline
-
-# Check if CUDA (GPU support) is available, for faster processing if you have a GPU
-device = 0 if torch.cuda.is_available() else -1
+import torch
 
 # Load the BART model for summarization using Hugging Face's pipeline
-summarizer = pipeline("summarization", model="facebook/bart-large-cnn", device=device)
+summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
 
-# Function to summarize a given text
+# Prediction function
 def summarize_text(input_text):
-    # Get the summary from the model
+    # Check if the input text is empty
+    if not input_text.strip():
+        return {"Error": "Please input some text"}
+    
+    # Use the summarizer to generate the summary
     summary = summarizer(input_text, max_length=150, min_length=50, do_sample=False)
+    
     return summary[0]['summary_text']
 
-# Example text to summarize
-text_to_summarize = """
-The Hugging Face team has developed several transformers architectures that excel in a variety of NLP tasks.
-These architectures are pre-trained on large text corpora and can be fine-tuned for specific tasks like text classification, named entity recognition, and question answering.
-The transformers library provides easy-to-use tools for downloading pre-trained models and fine-tuning them, making it simple for developers and researchers to apply state-of-the-art NLP methods to their own tasks.
-Hugging Face has made it easier for people to contribute to the world of NLP research by providing open-source libraries that are well-documented and easy to use.
-"""
-
-# Calling the function to summarize the text
-summary = summarize_text(text_to_summarize)
-
-# Print the summarized text
-print("Original Text: \n", text_to_summarize)
-print("\nSummary: \n", summary)
+# Gradio interface
+demo = gr.Interface(
+    fn=summarize_text, 
+    inputs=gr.Textbox(label="Enter Text to Summarize", lines=5),
+    outputs=gr.Label(label="Summary"), 
+    title="Text Summarizer using BART",
+    description="This app uses the `facebook/bart-large-cnn` model to summarize the text you input.",
+    allow_flagging="never"
+)
 
 # Launch the interface
 demo.launch()
+